Retinoids have classically been associated with the differentiation of cells of epithelioid origin and with pattern formation during development. Consistent with this, preliminary results demonstrate a relationship among retinoid-induced differentiation, growth inhibition and the functional expression of components of the cadherin-based adhesion and signalling system in breast cancer cells. Loss of E-cadherin and/or defects in cadherin-associated molecules have been implicated in breast tumor progression. It is one goal of the proposed work to clarify the role of cadherin based adhesion and signalling in retinoid action. We have identified the cadherin-associated molecule beta-catenin as being a key element in the breast cancer cell response to retinoids. Beta- catenin is thought to function not only in cadherin-based cell-cell adhesion but also in intracellular signalling. Preliminary results demonstrate that beta-catenin is a substrate for both tyrosine and serine/threonine kinases and that beta-catenin exists as a complex with a serine/threonine kinase and the tumor suppressor gene APC. Another goal of this proposal is to investigate the role of serine kinase activity in the response of breast cancer cells to retinoids. Retinoids and antiestrogens such as tamoxifen have also been proposed as chemopreventive agents in breast cancer. Preliminary results in one animal model of mammary carcinogenesis demonstrate for the first time that 9-cis retinoic acid, the natural ligand for the RXR class of retinoid receptors, can significantly decrease tumor incidence and tumor burden. Importantly, in combination with tamoxifen, 9-cis retinoic acid significantly extends tumor latency, further decreases tumor incidence and number, and increases the number of tumor free animals. An important goal of the proposed work will be To use a number of other animal models of mammary carcinogenesis to confirm the chemopreventive effects of 9-cis retinoic acid and synergy with tamoxifen. We plan to optimize chemoprevention strategies in these animal models prior to initiating combined retinoid and tamoxifen chemoprevention trials in humans. Although this proposal emphasizes the utility of retinoids in chemoprevention, retinoid treatment inhibits the growth of many overtly malignant breast cancer cells. Taken together with the ability of retinoid treatment to increase cadherin-dependent cell-cell adhesion thee data suggest that retinoids may be useful, not only in the prevention of breast cancer, but also in its treatment Our final goal is to use material from the phase 2 clinical trial to identify molecular and cellular markers which are associated with, and which may predIct responsiveness to, retinoid and combined retinoid/tamoxifen treatment.